The clinical literature has emphasized that the abuse potential of cocaine is related to its speed of onset and that each time a person experiences a rapid and intense cocaine rush there is an increased risk of further drug taking and an increased likelihood of addition. While it is well known that cocaine intake in rats shows a fast- rising loading phase, most rodent models of cocaine addiction instead have focused on the maintenance phase and have explored the effects of long access sessions. The general assumption is that more drug exposure and high intake produces an addicted phenotype. This grant challenges that premise. Our hypothesis is that brief episodes (eg. 5 min) of intense drug use are sufficient to cause a transition from recreational to binge-like patterns of intake. Our data show that self-administration procedures that engender spiking drug levels produce a more robust escalation of drug intake and a dramatic increase in the motivation to self- administer cocaine. Our hypothesis is that the number of 'spikes' (or loading phases) have a much greater impact on the addiction process than the maintenance phase or total drug intake. The experiments proposed in this grant are designed to confirm, extend and validate our initial findings. Specific Aim 1 will test the hypothesis that the number of spikes, the change in spike height and the rise time of each spike is an important determinant of escalation of drug intake and the motivation to response (as measured by a progressive ratio schedule). Specific Aim 1 will also test the hypothesis that cocaine self-administration is regulated by an endogenous circadian influence and that spiking cocaine levels can dysregulate this important physiological control mechanism. Our theoretical viewpoint draws heavily on modeling of cocaine concentrations in brain and it becomes important for us to examine and validate the assumptions underlying the model. Specific Aim 2 will validate the kinetic model using self-administration procedures and will examine the effects of cocaine consumption on extracellular cocaine and dopamine parameters. A method for separating appetitive and consummatory responding will be developed in Specific Aim 3. A two lever procedure will allows us to study cocaine consumption on one lever and the motivation to gain access to cocaine on the other lever. By using a PR schedule and manipulating timeouts and time of day the procedure will enable us to test hypotheses regarding the relationship between brain levels cocaine, drug seeking and drug taking.